Is ml agents dead?

In github, I see last commits like 4 months ago.

 

I was wondering the same thing. Can someone from the dev team comment the roadmap for ML Agents?

 

++ same question, we need some information about ml-agents

 

We are not allowed to ask questions in repo, yet nobody is checking here.

 

Truly hoping it's because of an IL2cpp integration/major refactor

 

So in general ML agents is living on and will get updates (in terms of new features and bugfixes, etc)?

 

Hey, glad to hear this! Any more concrete idea on when the State of the Union will take place? Thanks

 

My heart sank when news broke about the recent Unity layoffs and how it primarily affected AI and Engineering divisions. Even though I'm not actively using it right now, ml-agents had always been a joy to work with for me in the past. Hope it'll come out of the current predicament unscathed.

Btw, the cert on unity3d.ai has expired (this is the URL used on ml-agents' Github home page).

 

Same thoughts here.

I see great potential for A.I. like in every possible areas, including gaming. And ml-agents is just super nice to use. One of the main reasons I chose Unity over UE for my project.

 

thank you! i was really worried that Unity was to abandon ml-agents and in general all these PRICELESS open source projects.. some good friends have been laid off last months and was already looking for alternatives (Godot anyone?

but it ml-agents stays alive, here i remain. thank you!

 

How I read this between the lines is that people behind original ML-Agents team are not working on this anymore or am I understanding this wrong? How many people are working on this now?

Why not just post a blog post about current state and make a productboard for it on unity.com/roadmap? I don't get why anyone would want zoom meeting over that but maybe I'm just too old fashioned to understand this.

 

Hello,

July's almost over but I don't see any updates on the subject that we were promised.. Is it still on?

 

Looking forward to State Of The Union ML-Agents. August at the latest?

Unity Share price has been in freefall since november 2021. Peaked at 189, 5y average at about 100 $/share. Currently trading at 36. They have invested over a billion in takeovers and are trying to get into the VFX industry, thats why we see all the focus on cinemachine recently.

According to https://www.fool.com/investing/2022/07/07/why-unity-software-stock-plunged-74-in-the-first-h/ Q1 report had unsuspected 110 mil loss in revenue and the company will take steps in Q2 and Q3 to correct the problem.

In the last week of june we've heard that 4% of the company (about 200 people) have been layed off. Mostly the AI department and engineering. Looking at git commits https://github.com/Unity-Technologies/ml-agents/graphs/commit-activity it does seem that the entire ML agents team is gone or is moved to a private entity.

Best case scenario in my opinion:

A tiny team is still working on ML agents in private repos. In Q4 the shareprice skyrockets and the management decides to re-invest Ai and the development continues as it was with a delay of a year.

Worst case scanario:

Investments and VFX is a dud, company continues to bleed money, CEO is pumping his chest while playing monopoly. Financial crisis crushes stock prices even further. Unity will cut 60% of it's workforce ( except marketing and sales ) and will focus on strictly core features. All non money making R&D projects will be crucified and hung from the windows of the corporate office. ML agents will be the prime example of "wasted resources"

Update:

Just found a random comment on youtube:

Unity ML-Agents API are based on the Tensorflow library from Google, and Google is moving their ML division to use another library called Jax. Maybe Unity is also going in the same direction, and that explains some of the layoff in the AI department, or this could be only John Riccitielo invese king Midas power working again.

The CEO of unity technologies John Riccitiello has been CEO of EA when it was voted worst company of the year, twice in a row. He also said in an interview: if you're not thinking about monetisation during your creative process, you're a “F***ing idiot.”

It's a business world where billions are on the line, so maybe you need a sleek business man like that to run a corporation. On the other hand, the results are not there. Going from 190 to 40 dollar a share. 40 is half of what the company was worth before going public. Cutting engineering to save a buck. Sounds like a CEO from Wish

On 13th of july unity is cutting Gogaya project (ultimate AAA sample project tutorial)

Seems that all non money making initiatives are getting an axe one by one.

 

Damn!! ((
But why then all those promising messages above, from Unity stuff. Lies?

 

The messages are fairly vague. "We need to communicate better". "There will be a roadmap, zoom call with questions". It's the first step in a negotiation technique, self audit, showing empathy to diffuse frustrations ( check the book Never split the difference, by Christopher Voss )

I've worked at 2 companies that were stripped and sold off. I also had the pleasure developing in Shiva3D which ran out of money. Some similarities here. When things are bad, there is a big reachout to community/employees. Everyone talks, people can share opinions. Sometimes a creative board is formed, where representatives can brainstorm about future, improvements, rebranding etc. Then, after a little while: SOLD, CLOSED, FIRED, drop the keys here, clean the desk by 10am. No feedback, management is unreachable. Bye. And then it picks up again in 6-12 month.

 

I'm definitely worried about investing more time in building out my Unity ML product while Unity are killing unprofitable products and their CEO is ranting about monetisation.

The promise of 'better communication' followed by 100% radio silence isn't doing much to make me less worried.

 

Yes. The worst part is this silly silence. Just let us freaking know - bad news or good news its is. So we can plan our future work accordingly.

How hard is it to jump in forum and post a message of truth? Or even create a blog entry on the subject?

This guy "AndrewKemendoWork" seems to have serious memory problems. Giving promises and then just vanishing.

 

You got to cut those guys some slack... With a company the size of Unity, it'll probably take some back and forth until the dust settles and anybody knows anything. I'm sure the team lead would let us know if there was a definitive roadmap for ML-Agents.
Also, it's not going anywhere. The current release works pretty well and I don't see a good reason for not using it any longer, even if I knew development is on hold for the time being.

 

To be fair -- I don't think he's upset at the devs directly. Clearly Unity is having a hard time figuring out what to do internally, and it is frustrating as a developer when a company you may be trusting your livelihood on begins to hem and haw on what it will or won't support, and can't even give you a clear idea of what it's current thinking is. This is general anxiety -- not a direct attack on the Unity devs.
This anxiety is part of the "dust settling" process. Unity's direction (and silence) could affect his livelihood (and will affect the livelihood of many others). Maybe someone important at Unity will hear these fears and frustrations and try to ease them by offering people something more concrete. Or maybe Unity will just implode first because nobody there listens. :/

This is a bit of a reach. The team lead who promised to let us know something might have been canned or just relocated to some other (unrelated) team. Again, it's part of the "dust-settling" process. Expressing one's feelings and opinions is very much a necessary 'evil' to ensure those reading know that there is something wrong. Expressing passion (through fair arguments) matters in the enlightening process. It is on Unity to address them (or get what's coming for ignoring the dangers inherent in not addressing the reasonable arguments). Ignoring people tends to look waaay worse for a company than simply saying "I don't know -- but I will try to find out". But that part is just my opinion. Silence, however, isn't a great option when you don't know what to do. People expect to be able to trust companies to be genuine and upfront. They very often aren't -- but that's a deeper problem with Capitalism, and out of the scope of my reply.

That being said --

This part is definitely true. At the very least, one can work with what they have. But in a field where things are moving at a breakneck speed, it is reasonable to expect Unity to pitch-in and work with advancing this field too. But again - that is just my own opinion.

 

To give credit where credit is due - the old ML-Agents (and Barracuda) team created a really incredible product. This is speaking as a developer who's fully productionized ML-Agents in a game. Granted it's rough around the edges in some regards and needed lots of work for efficiency/overhead, no one else has provided a product like this, that can be easily deployed in a game, yet. If you spend anytime in the repo (C# and Python) you'd see that it wasn't an easy feat to achieve.

It's sad to know that a lot of talented people were let go, but as is, the product is still intact and is in a good place for others to use and for a new team to jump in and keep working/developing in.

I wish the old team the best and hope that we can turn the outlook of ML-Agents into something hopeful and less negative. Even with the old team gone, we are a community that can continue to innovate, share insights and developments, and provide useful feedback for future tech/improvements.

 

This is true. Looking into this code myself and, personally, I'm very grateful for their efforts!

I agree with you on this for the most part.

I feel like if we make any improvements on the "rough around the edges" parts, we should all give back to the memory of the team that made this possible by sharing our changes / upgrades to the process on github, provide documentation, or just give back in some way to make others' lives a little easier, so that we won't all have to keep reinventing the wheel so much. But that's just me...

 

http://www.siggraph2022.unity.com/
They have one session about AI and Physics Assisted Character Pose Authoring on Wednesday 5:10pm which is using ML agents....

 

The GitHub repo has not been updated with the latest changes. The last commit on the public repo was in January, but there was a new version released on PyPI in April that includes code changes that have not been put onto Git. Not sure if this was an oversight or what, but since it is still under the Apache License, I have copied the updates to my own fork so we can continue to maintain it if the Unity devs don't want to.

 

So, any news?

 

I don't think so. And I also don't think there ever will be either. Not unless Unity changes direction as a company - and fast.

Unity, as a company, seems to be in its death throes right now. When your whole Visual Scripting (and most other) projects go dark, instead of developing your core product further, you decide to diversify your income, and then, when that fails, you decide to sell your soul (i.e. a merger) to a subpar advertising agency responsible for the "high software standards" required by malware corporatization, you and your company are going through some serious crap. You don't even care how your company looks anymore - as long as it works.

But it won't. And it never does.

I'm sorry, but this will probably be dead until either Unity's CEO comes to his senses, or (more likely) things fail badly and he is eventually replaced. There is a reason EA won the golden swirl trophy under his tenure in 2012 (for being the worst company in America).

 

Now this

https://gamefromscratch.com/applovin-offers-17-5b-for-unity/

 

Ouch. Can you imagine making games in a game engine owned by a company called "AppLovin" ?

:S

He will "thoroughly review" the offer. And probably make sweet love to it at night too. Because it's going to save his bacon.


Damn. Guess we have to get ready to forget about stuff like ML agents. Unity is about to downsize. :/

 

Does anyone know a good alternative to ML-Agents for quickly implementing reinforcement learning in a game? I started using Unity, and ML-Agents, because I was hoping to make a game with machine learning-based AI as a core mechanic. If Unity is abandoning it, then I should switch platforms. Any recommendations?

 

Two of the individuals giving that presentation at SIGGRAPH
were Senior ML developers at Unity. According to LinkedIn, they are both still employed at Unity. Sliver of hope.

 

Yeah! Fingers crossed.

 

This SIGGRAPH presentation allegedly already happened and was scheduled for 11 minutes. Does anyone know how it went?

 

Ok August ends, we are w8ing))

 

Andrew hasn't browsed the forums since June 27 - anyone else from Unity want to step up and promise us a roadmap

 

He probably got the axe too. :/

 

Just found this thread, and it seems f*cked. Oo man, I was about to start learning ML agent, but its kind of risk for the long-term learning investment

 

Today I found this:

"Well done to our ML Artistry team who won the Audience Choice Award at ACM SIGGRAPH 2022 Real-Time Live!

Presented by Machine Learning Manager, Florent Bocquelet, “AI and Physics Assisted Character Pose Authoring” won the audience vote for their new research project around character animation and posing using machine learning. Unity’s character pose tool showed how users can work with complex, handcrafted constraints even with minimal artistic experience.

We’re very proud of the team for this accomplishment, and congratulations to all the other speakers for their fascinating presentations"

So this is most likely the part that is left from A.I. stuff. And these ML agents is R.I.P

 

Yes and no.

For me it was the perfect way to get started in reinforcement learning. Because of the easiness of setup and well documented hyperparameter tuning. And the huge amount of examples of course.

When it comes to reward assignment, what I think is most important part of RL, you wont risk much. Same strategies you can apply in any framework. Even if you go straight to Stable Baselines, etc.

And the current product is still pretty stable. You can create some vision of yours for sure. Maybe that is what this project needs in order to be resurrected -> attention - projects build on it - preferably commercial, so cash flows in to Unity

 

Here's a related paper by some of the same authors
https://arxiv.org/pdf/2208.08274.pdf
none of whom I recognize as being part of the previous ML-Agents team.

It's an interesting idea: estimating character poses from reference images and generating ml-aided IK for them. I can see how this could be rolled out as a productivity tool. However, it doesn't really look like a reinforcement learning problem you would solve with ML-Agents, and I don't expect there to be much overlap with the latter.

Besides, Nvidia seems to be ahead when it comes to creating physics-based character animations with reinforcement learning:

 

...hmmm, ML Agents is dead? I didn't read anything to promising. I ended up here when I found the lack of documentation on setup and configuration. Any thoughts?

 

I am very happy to hear that. I would also kindly ask to be more responsible to the community and check these forums even if it's twice per month. You would spare as all these debates and false assumptions.

 

Thanks for the update

 

Great, glad to hear it! Thanks for the update.

 

I hate to be the person saying this but this still follows the old pattern "but we have these updates coming up but only some day in the future" after a long radio silence. Unity has track record on these things and it's not pretty, wouldn't get hopes too up. We might eventually get some last half baked release and then people move on. It's pretty clear you just don't fire your old team that did amazing job on ML and then replace them with another team if the focus would still be the same. Things like that just don't happen and there's nothing here that indicates otherwise.

 

Lots of tickets closed on the github just now - good sign.

 

Speaking to alternatives, has anyone been able to accomplish handshaking between PyTorch and Unity like this article suggests.
https://medium.com/@a.abelhopereira/how-to-use-pytorch-models-in-unity-aa1e964d3374

 

Does anyone know if/when Unity ML agent updates will be released?